Methods and apparatuses for predicting risk of prostate cancer and prostate gland volume

1. A method performed using an assay system, the assay system comprising:

a plurality of analysis regions comprising binding partners adapted to bind multiple blood markers, the multiple blood markers comprising total prostate-specific antigen (tPSA), free prostate-specific antigen (fPSA), intact prostate-specific antigen (iPSA), and human kallikrein 2 (hK2), the plurality of analysis regions comprising:

a first analysis region comprising a first binding partner adapted to bind tPSA,

a second analysis region comprising a second binding partner adapted to bind fPSA,

a third analysis region comprising a third binding partner adapted to bind iPSA, and

a fourth analysis region comprising a fourth binding partner adapted to bind hK2;

multiple detection devices configured to detect presence of analytes in the plurality of analysis regions, wherein the multiple detection devices are configured to detect, using luminescent detection, emissions from the plurality of analysis regions indicative of the analytes bound to their respective binding partners, and wherein each of the analytes is selected from a group consisting of tPSA, fPSA, iPSA, and hK2,

the method comprising:

using a computer to perform:

obtaining a value for an age of a person associated with a sample;

receiving, from the multiple detection devices, values for tPSA and one or more of fPSA, iPSA, and hK2;

processing the value for the age and the value for tPSA to determine a probability of a positive biopsy for prostate cancer, wherein processing the value for the age and the value for tPSA comprises:

determining which logistic regression model of a plurality of logistic regression models to use for determining the probability of a positive biopsy for prostate cancer based on the value for tPSA received from the multiple detection devices, wherein determining which logistic regression model to use comprises:

selecting a first logistic regression model from the plurality of logistic regression models when the value of tPSA is below a threshold, the first logistic regression model comprising a first plurality of predictive variables corresponding to tPSA and one or more of fPSA, iPSA, and hK2, respectively; and

selecting a second logistic regression model from the plurality of logistic regression models when the value of tPSA is above the threshold, the second logistic regression model comprising a second plurality of predictive variables including a predictive variable corresponding to tPSA, wherein the first plurality of predictive variables is different from the second plurality of predictive variables;

after selecting the first logistic regression model or the second logistic regression model:

determining whether the selected logistic regression model includes a variable corresponding to each of the multiple blood markers;

when it is determined that the selected logistic regression model includes a variable corresponding to each of the multiple blood markers, evaluating the selected logistic regression model by scaling values of variables corresponding to the age and each of the multiple blood markers by their respective coefficient values to produce respective scaled values and summing the respective scaled values to determine a first logit; and

when it is determined that the selected logistic regression model does not include a variable corresponding to each of the multiple blood markers, evaluating the selected logistic regression model by scaling values of variables corresponding to the age and tPSA by their respective coefficient values to produce respective scaled values and summing the respective scaled values to determine a second logit; and

determining the probability of a positive biopsy for prostate cancer using the first logit or the second logit; and

outputting an indication of the determined probability of a positive biopsy for prostate cancer; and

performing a biopsy on the person associated with the sample based on the indication of the determined probability of a positive biopsy for prostate cancer.

2. The method performed using the assay system of claim 1 , wherein processing the value for the age and the value for tPSA further comprises:

defining a first spline for fPSA;

calculating a first fPSA value by evaluating the first spline for fPSA using the value for fPSA;

defining a first spline for tPSA; and

calculating a first tPSA value by evaluating the first spline for tPSA using the value for tPSA,

wherein the values of variables corresponding to fPSA and tPSA, each of which is scaled by the respective coefficient value, further includes the calculated first fPSA value and the calculated first tPSA value.

3. The method performed using the assay system of claim 2 , wherein processing the value for the age and the value for tPSA further comprises:

defining a second spline for fPSA;

calculating a second fPSA value by evaluating the second spline for fPSA using the value for fPSA;

defining a second spline for tPSA; and

calculating a second tPSA value by evaluating the second spline for tPSA using the value for tPSA,

wherein the values of variables corresponding to fPSA and tPSA, each of which is scaled by the respective coefficient value, further includes the calculated second fPSA value and the calculated second tPSA value.

4. The method performed using the assay system of claim 3 , wherein each of the first spline for fPSA, the second spline for fPSA, the first spline for tPSA, and the second spline for tPSA is a cubic spline.

5. The method performed using the assay system of claim 1 , wherein a technique utilized for the luminescent detection is selected from the group consisting of photoluminescence, fluorescence, chemiluminescence, bioluminescence, and electrochemiluminescence.

6. The method performed using the assay system of claim 1 , wherein the multiple detection devices are configured to detect chemiluminescent or fluorescent emissions, and wherein the chemiluminescent or fluorescent emissions are indicative of the analytes bound to their respective binding partners.

7. The method performed using the assay system of claim 1 , wherein at least one of the multiple detection devices detects variation of a light signal as a function of time.

8. The method performed using the assay system of claim 1 , wherein the sample is whole blood, serum, or plasma.

9. The method performed using the assay system of claim 1 , further comprising a light source.

10. The method performed using the assay system of claim 1 , wherein the first binding partner and the second binding partner are different.

11. The method performed using the assay system of claim 10 , wherein:

the first logistic regression model comprises a respective coefficient value for the age and fPSA, iPSA, tPSA, and hK2 variables, and the second logistic regression model comprises a respective coefficient value for the age and tPSA variables.

12. The method performed using the assay system of claim 11 , wherein selecting the first logistic regression model or the second logistic regression model further comprises:

selecting the first logistic regression model when the value of tPSA is <15 ng/ml; and

selecting the second logistic regression model when the value of tPSA is ≥15 ng/ml.

13. The method performed using the assay system of claim 1 , wherein the indication of the probability of a positive biopsy for prostate cancer comprises an interpretable scale that is used to a guide a decision on whether the person obtains a biopsy.

14. The method performed using the assay system of claim 1 , wherein the indication of the probability of a positive biopsy for prostate cancer comprises a determination of a positive biopsy for a high-grade cancer or a low-grade cancer.

15. A method, comprising:

obtaining a value for an age of a person associated with a sample;

detecting, by multiple detection devices, presence of analytes in a plurality of analysis regions comprising binding partners adapted to bind multiple blood markers, the multiple blood markers comprising total prostate-specific antigen (tPSA), free prostate-specific antigen (fPSA), intact prostate-specific antigen (iPSA), and human kallikrein 2 (hK2), wherein:

the multiple detection devices are configured to detect, using luminescent detection, emissions from the plurality of analysis regions indicative of the analytes bound to their respective binding partners,

the plurality of analysis regions comprises a first analysis region comprising a first binding partner adapted to bind tPSA, a second analysis region comprising a second binding partner adapted to bind fPSA, a third analysis region comprising a third binding partner adapted to bind iPSA, and a fourth analysis region comprising a fourth binding partner adapted to bind hK2y, and

each of the analytes is selected from a group consisting of tPSA, fPSA, iPSA, and hK2;

processing the value for the age and value of tPSA to determine a probability of a positive biopsy for prostate cancer, wherein processing the value for the age and the value of tPSA comprises:

determining which logistic regression model of a plurality of logistic regression models to use for determining the probability of a positive biopsy for prostate cancer based on the value of tPSA received from the multiple detection devices, wherein determining which logistic regression model to use comprises:

selecting a first logistic regression model from the plurality of logistic regression models when the value of tPSA is below a threshold, the first logistic regression model comprising a first plurality of predictive variables corresponding to tPSA and one or more of fPSA, iPSA, and hK2, respectively; and

selecting a second logistic regression model from the plurality of logistic regression models when the value of tPSA is above the threshold, the second logistic regression model comprising a second plurality of predictive variables including a predictive variable corresponding to tPSA, wherein the first plurality of predictive variables is different from the second plurality of predictive variables;

after selecting the first logistic regression model or the second logistic regression model:

determining whether the selected logistic regression model includes a variable corresponding to each of the multiple blood markers;

when it is determined that the selected logistic regression model includes a variable corresponding to each of the multiple blood markers, evaluating the selected logistic regression model by scaling, by at least one computer processor, values of variables corresponding to the age and each of the multiple markers by their respective coefficient values to produce respective scaled values and summing the respective scaled values to determine a first logit; and

when it is determined that the selected logistic regression model does not include a variable corresponding to each of the multiple blood markers, evaluating the selected logistic regression model by scaling values of variables corresponding to the age and tPSA by their respective coefficient values to produce respective scaled values and summing the respective scaled values to determine a second logit; and

determining the probability of a positive biopsy for prostate cancer using the first logit or the second logit;

outputting an indication of the determined probability of a positive biopsy for prostate cancer; and

performing a biopsy on the person associated with the sample based on the indication of the determined probability of a positive biopsy for prostate cancer.

16. A method performed using an assay system, the assay system comprising:

a plurality of analysis regions comprising binding partners adapted to bind multiple blood markers, the multiple blood markers comprising total prostate-specific antigen (tPSA), free prostate-specific antigen (fPSA), intact prostate-specific antigen (iPSA), and human kallikrein 2 (hK2), the plurality of analysis regions comprising:

a first analysis region comprising a first binding partner adapted to bind tPSA,

a second analysis region comprising a second binding partner adapted to bind fPSA,

a third analysis region comprising a third binding partner adapted to bind iPSA, and

a fourth analysis region comprising a fourth binding partner adapted to bind hK 2 ;

multiple detection devices configured to detect presence of analytes in the plurality of analysis regions, wherein:

the multiple detection devices are configured to detect, using luminescent detection, emissions from the plurality of analysis regions indicative of the analytes bound to their respective binding partners, and

each of the analytes is selected from a group consisting of tPSA, fPSA, iPSA, and hK2,

the method comprising:

using a computer to perform:

obtaining a value for an age of a person associated with a sample;

determining a probability of a positive biopsy for prostate cancer, wherein determining the probability comprises:

determining which logistic regression model of a plurality of logistic regression models to use for determining the probability of a positive biopsy for prostate cancer based on a value of tPSA received from the multiple detection devices, wherein determining which logistic regression model to use comprises:

 selecting a first logistic regression model from the plurality of logistic regression models when the value of tPSA is below a threshold, the first logistic regression model comprising a first plurality of predictive variables corresponding to tPSA and one or more of fPSA, iPSA, and hK2, respectively; and

 selecting a second logistic regression model from the plurality of logistic regression models when the value of tPSA is above the threshold, the second logistic regression model comprising a second plurality of predictive variables including a predictive variable corresponding to tPSA, wherein the first plurality of predictive variables is different from the second plurality of predictive variables;

after selecting the first logistic regression model or the second logistic regression model:

 determining whether the selected logistic regression model includes a variable corresponding to each of the multiple blood markers;

 when it is determined that the selected logistic regression model includes a variable corresponding to each of the multiple blood markers, evaluating the selected logistic regression model by scaling values of variables corresponding to the age and each of the multiple blood markers by their respective coefficient values to produce respective scaled values and summing the respective scaled values to determine a first logit; and

 when it is determined that the selected logistic regression model does not include a variable corresponding to each of the multiple blood markers, evaluating the selected logistic regression model by scaling values of variables corresponding to the age and tPSA by their respective coefficient values to produce respective scaled values and summing the respective scaled values to determine a second logit; and

determining the probability of a positive biopsy for prostate cancer using the first or the second logit; and

outputting an indication of the determined probability of a positive biopsy for prostate cancer; and

performing a biopsy on the person associated with the sample based on the indication of the determined probability of a positive biopsy for prostate cancer.